Method and system for determining steady state of low-temperature adiabatic cylinder static evaporation rate test

Abstract

The application discloses a kind of low-temperature adiabatic gas cylinder static evaporation rate test steady-state determination method and system, and relates to low-temperature adiabatic gas cylinder detection technical field.Low-temperature adiabatic gas cylinder is placed after being filled with low-temperature medium, and with preset sampling period, the pressure data in the gas cylinder, gas phase space temperature data and environmental temperature data are continuously collected;Calculate the pressure change in unit time, form the pressure growth rate sequence;Calculate the gas phase temperature change in unit time, form the gas phase temperature change rate sequence;Calculate the environmental temperature change in unit time, form the environmental temperature change rate sequence;Parallel sliding window analysis is carried out on the pressure growth rate sequence, whether the fluctuation of pressure growth rate in continuous time window tends to converge is judged;When pressure stability condition, temperature stability condition and trajectory convergence condition are satisfied simultaneously, it is judged that low-temperature adiabatic gas cylinder enters steady-state stage.Can shorten test preparation time, reduce the risk of misjudgment, improve test efficiency.

Description

Technical Field

[0001] This invention relates to the field of cryogenic insulated gas cylinder testing technology, and in particular to a method and system for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders. Background Technology

[0002] Cryogenic insulated gas cylinders are refillable, mobile special pressure vessels primarily used for storing and transporting cryogenic media. Cryogenic media refer to substances that provide a sub-room temperature environment through heat transfer; these are liquefied gases or fluids used in refrigeration or cryogenic processes to obtain and maintain a cryogenic state, such as liquid nitrogen, liquid oxygen, liquid argon, liquid hydrogen, and liquefied natural gas (LNG). The insulation performance of cryogenic insulated gas cylinders directly affects the storage time and transportation safety of cryogenic liquids. Static evaporation rate is the most important technical indicator for measuring the insulation performance of cryogenic insulated gas cylinders. It refers to the percentage of gas mass evaporated due to external heat leakage within 24 hours after the cylinder reaches thermal equilibrium at its rated fill rate.

[0003] In static evaporation rate testing, to ensure the accuracy and repeatability of the test results, the gas cylinder usually needs to be left to stand for a relatively long period after filling to allow it to reach thermal equilibrium, i.e., enter the steady-state phase. Currently, a fixed-time standing period is commonly used as the test start condition, requiring 48 hours of standing time, which has the following drawbacks: 1. A fixed settling time lacks adaptability to differences in cylinder specifications, media types, and insulation structures. Larger cylinders (e.g., over 1000L) require a longer time to reach thermal equilibrium, while smaller cylinders (e.g., under 100L) reach steady state more quickly. Similarly, the thermal equilibrium time differs significantly between high-vacuum multilayer insulated cylinders and powder-insulated cylinders. Adopting a uniform 48-hour settling time leads to wasted time or insufficient preparation.

[0004] 2. Under certain operating conditions, gas cylinders reach a steady state much faster than the fixed time (e.g., the thermal equilibrium rate is accelerated when the ambient temperature is high in summer), resulting in wasted test preparation time and low test efficiency. According to statistics, using the fixed 48-hour settling method, about 60% of gas cylinders reach a steady state within 36 hours, and about 30% reach a steady state within 24 hours.

[0005] 3. Judging solely by time makes it difficult to identify the impact of abnormal factors such as changes in ambient temperature and operational disturbances on the formation of steady state. For example, if a large fluctuation in ambient temperature, an accidental collision with the gas cylinder, or strong airflow disturbance occurs during the settling process, even after 48 hours of settling, a true steady state may not have been reached. Conversely, if the settling process is conducted in an ideal environment, stability may be achieved in as little as 30 hours.

[0006] 4. Test initiation is highly dependent on human experience; different testers may have differing subjective judgments about whether the test has stabilized, leading to insufficient consistency in results. This is especially true in batch testing, where differences in human judgment introduce additional measurement uncertainty.

[0007] Furthermore, there are some improvements in existing technologies. For example, by adjusting multiple emission flow rates and measuring the corresponding pressure change rates, a linear relationship can be fitted to solve for the evaporation flow rate. However, this requires changing the emission state of the gas cylinder, i.e., actively emitting gas, and the purpose is to quickly measure the evaporation rate itself, rather than determining when to start testing. Another approach is to use the ratio of flow difference to judge flow stability. This is based on flow data during the active emission process, and using only a single window ratio makes it difficult to accurately identify the continuous evolution process of the gas cylinder from unsteady to steady state, and it cannot cope with abnormal situations such as sudden changes in ambient temperature. Summary of the Invention

[0008] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a method and system for determining the steady-state condition of a cryogenic insulated gas cylinder for static evaporation rate testing. By synergistically analyzing the changes in multi-dimensional parameters such as internal pressure and temperature of the gas cylinder, the method automatically identifies the timing when the gas cylinder enters steady-state testing conditions, thereby shortening test preparation time, reducing the risk of misjudgment, and improving test efficiency while ensuring test accuracy.

[0009] On one hand, embodiments of the present invention provide a method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders, including: After the cryogenic medium is filled, the cryogenic insulated gas cylinder is left to stand, and the internal pressure data, gas phase space temperature data and ambient temperature data are continuously collected at a preset sampling period. Based on pressure data from adjacent sampling times, the pressure change per unit time is calculated to form a pressure growth rate sequence. Based on the gas phase space temperature data at adjacent sampling times, the change in gas phase temperature per unit time is calculated to form a sequence of gas phase temperature change rates. Based on the ambient temperature data at adjacent sampling times, the change in ambient temperature per unit time is calculated to form an ambient temperature change rate sequence. The pressure growth rate sequence is subjected to parallel sliding window analysis using short and long time windows to determine whether the fluctuation of the pressure growth rate tends to converge within the continuous time window. When the pressure stability condition, temperature stability condition, and trajectory convergence condition are met simultaneously, the cryogenic adiabatic cylinder is determined to have entered the steady state stage. The pressure stability condition is as follows: the fluctuation of the pressure growth rate tends to converge within a continuous time window, and the coefficient of variation of the pressure growth rate within the continuous time window is less than the adaptive steady-state determination threshold. The temperature stability condition is as follows: the absolute value of each value in the gas phase temperature change rate sequence is less than the first temperature change rate threshold for three or more consecutive time windows, and the absolute value of each value in the ambient temperature change rate sequence is less than the second temperature change rate threshold for three or more consecutive time windows. The trajectory convergence condition is as follows: the curvature of the evolution trajectory of the multidimensional feature vector composed of the pressure growth rate, the gas phase temperature change rate and the ambient temperature change rate in the feature space is less than a preset curvature threshold, and the trajectory offset is less than a preset offset threshold, for at least two consecutive time windows.

[0010] According to some embodiments of the present invention, the preset sampling period is adaptively set according to the gas cylinder volume, and the sampling period ranges from 1 to 60 seconds.

[0011] According to some embodiments of the present invention, the short time window is 5 to 30 minutes long, and the long time window is 30 to 120 minutes long; when the coefficient of variation of the pressure growth rate within the short time window is less than a first local threshold, and the coefficient of variation of the pressure growth rate within the long time window is less than a second local threshold, it is determined that the pressure growth rate tends to converge at both the local and global scales, wherein the first local threshold is less than the second local threshold.

[0012] According to some embodiments of the present invention, the generation of the adaptive steady-state determination threshold includes: Obtain the nominal volume, design pressure rating, cryogenic medium type, and insulation structure of the cryogenic insulated gas cylinder, and generate the basic steady-state judgment threshold. Based on historical data from K consecutive time windows preceding the current time window, a historical volatility factor is calculated. This historical volatility factor is equal to the ratio of the current sliding window standard deviation to the median of the historical sliding window standard deviations. The adaptive steady-state determination threshold is obtained by multiplying the basic steady-state determination threshold by the historical fluctuation amplitude factor and the preset safety factor, where the safety factor ranges from 0.8 to 1.2.

[0013] According to some embodiments of the present invention, the multidimensional feature vector is subjected to dimensionality reduction processing using principal component analysis, and the high-dimensional features are projected into a low-dimensional space before the trajectory curvature and trajectory offset are calculated.

[0014] According to some embodiments of the present invention, the formula for calculating the pressure change per unit time to form a pressure growth rate sequence is as follows: ; In the formula, r for The rate of pressure increase over time, expressed in kPa / h; for The pressure value at any moment, for The pressure value at any given moment; The sampling period is in seconds; 3600 is a time unit conversion factor, which converts the rate of change per second to the rate of change per hour. The formula for calculating the change in gas phase temperature per unit time to form a sequence of gas phase temperature change rates is as follows: ; In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is in seconds; 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute. The formula for calculating the change in ambient temperature per unit time, forming a sequence of ambient temperature change rates, is as follows: ; In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is in seconds; 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute. The formulas for calculating the mean, standard deviation, and coefficient of variation of the pressure increase rate within a continuous time window are as follows: ; In the formula, μ r r represents the average rate of pressure increase within the sliding window, expressed in kPa / h. j σ represents the j-th pressure increase rate within the sliding window, in kPa / h; n is the number of sampling points within the sliding window; r CV represents the standard deviation of the pressure increase rate within the sliding window, in kPa / h. r is the coefficient of variation of the pressure increase rate.

[0015] According to some embodiments of the present invention, after the step of determining that the cryogenic adiabatic cylinder has entered the steady-state stage, the method further includes: Verify whether the duration of the steady-state phase exceeds a preset minimum duration threshold, wherein the minimum duration threshold ranges from 0.5 to 2 hours; The test verifies whether the relative deviation between the average pressure growth rate during the steady-state phase and the average pressure growth rate in the previous phase is less than a preset verification deviation threshold, wherein the value of the verification deviation threshold ranges from 5% to 10%. If all the above verifications pass, the steady-state phase is confirmed to be effective, and a static evaporation rate test start signal is generated; otherwise, the steady-state evolution analysis continues until the verification passes.

[0016] According to some embodiments of the present invention, the method further includes: Identify the types of abnormal disturbances, including transient jump anomalies, continuous drift anomalies, and environmental step anomalies; Implement a tiered response based on the type of abnormal disturbance: For transient jump anomalies, the steady-state determination process is not paused; instead, a data repair algorithm is used to remove and replace the anomalies. For persistent drift anomalies, the steady-state determination process is paused, and the steady-state evolution analysis is re-executed after no persistent drift is detected for three or more consecutive time windows. For environmental step anomalies, the steady-state determination process is paused, the historical data cache of the sliding window is reset, and the steady-state evolution analysis is re-executed after the environmental temperature stabilizes again.

[0017] According to some embodiments of the present invention, the transient jump anomaly is defined as: the pressure data shows a sudden change exceeding a preset jump threshold at a single sampling point, and returns to normal within the subsequent 1 to 2 sampling periods; The continuous drift anomaly is defined as follows: the pressure growth rate shows a monotonically increasing or monotonically decreasing trend within multiple consecutive time windows, and the trend intensity exceeds the preset drift detection threshold. The environmental step anomaly is defined as: the ambient temperature changes beyond a preset environmental step threshold within a short period of time, and the magnitude of the change exceeds three times the standard deviation of the rate of change of ambient temperature.

[0018] In another aspect, embodiments of the present invention provide a steady-state determination system for static evaporation rate testing of cryogenic insulated gas cylinders, used to implement the aforementioned steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders. The system includes: The data acquisition module is used to allow the cryogenic insulated gas cylinder to stand after it has been filled with cryogenic medium, and to continuously collect internal pressure data, gas phase space temperature data and ambient temperature data of the gas cylinder at a preset sampling period. The feature parameter calculation module, connected to the data acquisition module, is used to calculate the pressure growth rate sequence based on the pressure data at adjacent sampling times, calculate the gas phase temperature change rate sequence based on the gas phase space temperature data at adjacent sampling times, and calculate the ambient temperature change rate sequence based on the ambient temperature data at adjacent sampling times. The multi-scale analysis module, connected to the feature parameter calculation module, is used to perform parallel sliding window analysis on the pressure growth rate sequence using short and long time windows to determine whether the fluctuation of the pressure growth rate tends to converge at both the local and global scales. The steady-state judgment module is connected to the feature parameter calculation module and the multi-scale analysis module respectively, and is used to determine whether the cryogenic adiabatic cylinder has entered the steady-state stage.

[0019] The method and system for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to embodiments of the present invention have at least the following beneficial effects: By analyzing the fluctuation convergence of the pressure growth rate through a multi-scale sliding window and constructing a multi-dimensional feature vector by combining the gas phase temperature change rate and the ambient temperature change rate, the steady-state platform stage is jointly determined from three dimensions: pressure stability, temperature stability, and trajectory convergence. This reduces the risk of misjudgment and improves the accuracy and intelligence of the judgment. It can automatically identify steady-state conditions and start the test within 12–18 hours, reducing liquid nitrogen consumption, shortening test preparation time, improving test efficiency, and resulting in significant economic benefits. The entire process is automatically collected, calculated, judged, and generates the test start signal without manual intervention; the high degree of automation reduces reliance on human experience and improves the objectivity and reproducibility of the test results. Through a three-level abnormal disturbance identification and graded response mechanism, instantaneous jump anomalies are repaired without interrupting the judgment; continuous drift anomalies are paused and resumed after recovery; and environmental step anomalies are reset and buffered until stabilization. It has strong anti-interference capabilities and good environmental adaptability, effectively avoiding misjudgments caused by interference such as air conditioning start-up and shutdown, personnel movement, and sensor noise, enabling the system to operate stably in real industrial environments.

[0020] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of the steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders according to an embodiment of the present invention; Figure 2This is a schematic diagram of the pressure-time variation curve of the steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the pressure change rate versus time curve of the steady-state determination method for static evaporation rate testing of low-temperature insulated gas cylinders according to an embodiment of the present invention. Figure 4 This is a schematic diagram of the pressure growth rate variation coefficient curve of the steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders according to an embodiment of the present invention. Figure 5 This is a schematic diagram of the evolution trajectory of the multidimensional feature vector in the feature space of the steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders according to an embodiment of the present invention. Figure 6 This is a schematic diagram comparing the steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders according to an embodiment of the present invention with the test results of traditional methods; Figure 7 This is a functional block diagram of the steady-state determination system for static evaporation rate testing of cryogenic insulated gas cylinders according to an embodiment of the present invention. Detailed Implementation

[0022] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0023] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0024] In the description of this invention, "several" means one or more, "multiple" means two or more, "greater than," "less than," "exceeding," etc. are understood to exclude the stated number, and "above," "below," "within," etc. are understood to include the stated number. If "first," "second," etc. are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of indicated technical features or the order of the indicated technical features.

[0025] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] This embodiment provides a steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders. Please refer to [link to relevant documentation]. Figure 1 The steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders mainly includes steps S101~S106: S101. After the cryogenic medium is filled, the cryogenic insulated gas cylinder is left to stand, and the internal pressure data, gas phase space temperature data and ambient temperature data are continuously collected at a preset sampling period.

[0027] S102. Based on the pressure data at adjacent sampling times, calculate the pressure change per unit time to form a pressure growth rate sequence.

[0028] S103. Based on the gas phase space temperature data at adjacent sampling times, calculate the change in gas phase temperature per unit time to form a gas phase temperature change rate sequence.

[0029] S104. Based on the ambient temperature data at adjacent sampling times, calculate the change in ambient temperature per unit time to form an ambient temperature change rate sequence.

[0030] S105. Parallel sliding window analysis is performed on the pressure growth rate sequence using short and long time windows to determine whether the fluctuation of the pressure growth rate tends to converge within the continuous time window.

[0031] S106. When the pressure stability condition, temperature stability condition, and trajectory convergence condition are simultaneously met, the cryogenic adiabatic cylinder is determined to have entered the steady-state stage. The pressure stability condition is: the fluctuation of the pressure growth rate tends to converge within a continuous time window, and the coefficient of variation of the pressure growth rate within the continuous time window is less than the adaptive steady-state determination threshold. The temperature stability condition is: the absolute value of each value in the gas phase temperature change rate sequence is less than the first temperature change rate threshold for three or more consecutive time windows, and the absolute value of each value in the ambient temperature change rate sequence is less than the second temperature change rate threshold for three or more consecutive time windows. The trajectory convergence condition is: the curvature of the evolution trajectory of the multidimensional feature vector composed of the pressure growth rate, gas phase temperature change rate, and ambient temperature change rate in the feature space is less than a preset curvature threshold, and the trajectory offset is less than a preset offset threshold, lasting for at least two consecutive time windows.

[0032] In step S101 above, the preset sampling period is adaptively set according to the gas cylinder volume, and the sampling period range is 1 to 60 seconds.

[0033] In step S102 above, the formula for calculating the pressure change per unit time to form the pressure growth rate sequence is as follows: ; In the formula, r for The rate of pressure increase over time, expressed in kPa / h; for The pressure value at any moment, for The pressure value at any given moment; The sampling period is in seconds; 3600 is a time unit conversion factor, which converts the rate of change per second to the rate of change per hour.

[0034] In step S103 above, the formula for calculating the gas phase temperature change rate sequence per unit time is as follows: ; In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is in seconds; 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute.

[0035] In step S104 above, the formula for calculating the change in ambient temperature per unit time and forming the ambient temperature change rate sequence is as follows: ; In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is in seconds; 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute.

[0036] In step S105 above, the length of the short time window is 5 to 30 minutes, and the length of the long time window is 30 to 120 minutes. When the coefficient of variation of the pressure growth rate within the short time window is less than the first local threshold, and the coefficient of variation of the pressure growth rate within the long time window is less than the second local threshold, it is determined that the pressure growth rate tends to converge on both the local and global scales, wherein the first local threshold is less than the second local threshold.

[0037] In step S105 above, the formulas for calculating the mean, standard deviation, and coefficient of variation of the pressure increase rate within a continuous time window are as follows: ; In the formula, μ r r represents the average rate of pressure increase within the sliding window, expressed in kPa / h. j σ represents the j-th pressure increase rate within the sliding window, in kPa / h; n is the number of sampling points within the sliding window; r CV represents the standard deviation of the pressure increase rate within the sliding window, in kPa / h. r is the coefficient of variation of the pressure increase rate.

[0038] In step S106 above, the generation of the adaptive steady-state determination threshold includes: Obtain the nominal volume, design pressure rating, cryogenic medium type, and insulation structure of the cryogenic insulated gas cylinder, and generate the basic steady-state judgment threshold. Based on historical data from K consecutive time windows prior to the current time window, calculate the historical volatility factor, which is equal to the ratio of the current sliding window standard deviation to the median of the historical sliding window standard deviations. The adaptive steady-state determination threshold is obtained by multiplying the basic steady-state determination threshold by the historical fluctuation amplitude factor and the preset safety factor, where the safety factor ranges from 0.8 to 1.2.

[0039] In step S106 above, the multidimensional feature vectors are reduced in dimensionality using principal component analysis. The high-dimensional features are projected into a low-dimensional space before the trajectory curvature and trajectory offset are calculated.

[0040] Following the step of determining whether the cryogenic adiabatic cylinder has entered the steady-state stage in step S106 above, the following is also included: Verify whether the duration of the steady-state phase exceeds a preset minimum duration threshold, which ranges from 0.5 to 2 hours. Verify whether the relative deviation between the average pressure growth rate during the steady-state phase and the average pressure growth rate in the previous phase is less than a preset verification deviation threshold, with the verification deviation threshold ranging from 5% to 10%. If all the above verifications pass, the steady-state phase is confirmed to be effective, and a static evaporation rate test start signal is generated; otherwise, the steady-state evolution analysis continues until the verification passes.

[0041] In some embodiments of the present invention, a multi-level abnormal disturbance identification and graded response step is also included: Identify the types of anomalous disturbances, including transient jump anomalies, persistent drift anomalies, and environmental step anomalies; Implement a tiered response based on the type of abnormal disturbance: For transient jump anomalies, the steady-state determination process is not paused; instead, a data repair algorithm is used to remove and replace the anomalies. For persistent drift anomalies, the steady-state determination process is paused, and the steady-state evolution analysis is re-executed after no persistent drift is detected for three or more consecutive time windows. For environmental step anomalies, the steady-state determination process is paused, the historical data cache of the sliding window is reset, and the steady-state evolution analysis is re-executed after the environmental temperature stabilizes again.

[0042] It should be noted that the instantaneous jump anomaly is defined as: a sudden change in pressure data at a single sampling point that exceeds the preset jump threshold, and then returns to normal within the next 1 to 2 sampling periods; the continuous drift anomaly is defined as: the pressure growth rate shows a monotonically increasing or monotonically decreasing trend over multiple consecutive time windows, and the trend intensity exceeds the preset drift detection threshold; the environmental step anomaly is defined as: the ambient temperature changes by more than the preset environmental step threshold in a short period of time, and the magnitude of the change exceeds three standard deviations from the rate of change of ambient temperature.

[0043] The following is a detailed description of the steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders provided in embodiments of the present invention: 1. Data Collection After the cryogenic medium is filled, the cryogenic insulated gas cylinder is left to stand, and internal pressure data, gas phase space temperature data, and ambient temperature data are continuously collected at a preset sampling period.

[0044] The preset sampling period is adaptively set according to the cylinder volume; the larger the volume, the longer the sampling period, ranging from 1 to 60 seconds. Specifically, for small cylinders with a volume ≤ 100L, the sampling period is 1–5 seconds; for medium-sized cylinders with a volume of 100L–500L, the sampling period is 10–30 seconds; and for large cylinders or storage tanks with a volume ≥ 500L, the sampling period is 30–60 seconds. This adaptive sampling strategy ensures data resolution while avoiding data redundancy and storage pressure.

[0045] Pressure data is measured using a high-precision pressure sensor with a range of 1.5 to 2 times the nominal working pressure of the gas cylinder and an accuracy of not less than 0.1%FS. Temperature data is measured using a platinum resistance temperature sensor (such as Pt100) with an accuracy of not less than ±0.1℃ and a response time constant of less than 1 second.

[0046] 2. Calculation of characteristic parameters Based on pressure data from adjacent sampling times, the pressure change per unit time is calculated to form a pressure growth rate sequence r(t): ; In the formula, r for The rate of pressure increase over time, expressed in kPa / h; for The pressure value at any moment, for The pressure value at any given moment; The sampling period is in seconds; 3600 is a time unit conversion factor, which converts the rate of change per second to the rate of change per hour.

[0047] Based on the gas phase space temperature data at adjacent sampling times, the change in gas phase temperature per unit time is calculated to form a gas phase temperature change rate sequence. : ; In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is in seconds; 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute.

[0048] Based on ambient temperature data from adjacent sampling times, the change in ambient temperature per unit time is calculated to form an ambient temperature change rate sequence. : ; In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is in seconds; 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute.

[0049] Calculate the local statistical characteristics of the pressure growth rate sequence. For each sliding time window, calculate the mean μ of the pressure growth rate within the window. r Standard deviation σ r and coefficient of variation (CV) r : ; In the formula, μ r r represents the average rate of pressure increase within the sliding window, expressed in kPa / h. jσ represents the j-th pressure increase rate within the sliding window, in kPa / h; n is the number of sampling points within the sliding window; r CV represents the standard deviation of the pressure increase rate within the sliding window, in kPa / h. r is the coefficient of variation of the pressure increase rate.

[0050] Coefficient of variation (CV) r It is a dimensionless measure of fluctuation, ranging from 0 to 0.3, and typically less than 0.08 in the steady state. It reflects the relative degree of fluctuation better than the standard deviation and is suitable for comparing pressure growth rates of different magnitudes.

[0051] 3. Multi-scale sliding window fluctuation convergence analysis The pressure growth rate sequence is analyzed in parallel using at least two time windows of different lengths, for example, using a short time window W. s and long window W L Short time windows are used to capture the immediate stability of local fluctuations, while long time windows are used to judge the convergence of the overall trend.

[0052] Specifically, short time window W s The length range is 5 to 30 minutes, for example, 10 minutes; long window W L The duration ranges from 30 to 120 minutes, for example, 60 minutes. The two windows advance in a sliding motion, with a sliding step size that can be set from 1 to 5 minutes.

[0053] The window length is adjusted according to the cylinder volume and testing accuracy requirements. For small cylinders with a volume less than 100L, the window length can be appropriately shortened (W). s =5 minutes, W L =30 minutes); for large storage tanks with a volume greater than 1000L, the window length can be extended (W s =30 minutes, W L =120 minutes). The window length is related to the thermal time constant of the gas cylinder, and is generally taken as 1 / 10 to 1 / 3 of the thermal time constant.

[0054] When the coefficient of variation (CV) of the pressure growth rate within a short time window r (W s () less than the first local threshold Furthermore, the coefficient of variation (CV) of the pressure growth rate within the long-term window r (W L () less than the second local threshold At that time, it was determined that the rate of pressure increase tended to converge on both the local and global scales.

[0055] The relationship between the first local threshold θ1 and the second local threshold θ2 is θ1 < θ2, meaning the local judgment condition is stricter than the overall judgment condition. θ1 ranges from 0.03 to 0.08, meaning local fluctuations do not exceed 3% to 8% of the mean; θ2 ranges from 0.05 to 0.12. This design logic of strict short-window and lenient long-window is based on the principle that local stability is the foundation of overall stability; only when fluctuations are sufficiently small in a short time can significant drift not occur over a long period.

[0056] 4. Adaptive steady-state determination threshold generation Obtain the nominal volume V and design pressure rating P of the cryogenic insulated gas cylinder. design Cryogenic medium type M type Insulation structure type I type These parameters can be obtained by querying the database.

[0057] Based on these parameters, a basic steady-state determination threshold θ is generated using a pre-trained regression model or an empirical lookup table. base For example, for a 450L liquid nitrogen high-vacuum multilayer insulated cylinder, θ base =0.08; For a 175L liquid nitrogen high-vacuum multilayer insulated gas cylinder, θ base =0.10; For a 450L liquid nitrogen powder insulated gas cylinder (with poor insulation performance), θ base =0.12. The influence of different media types: liquid hydrogen has a lower boiling point and a smaller latent heat of vaporization, so the pressure increase rate fluctuates more, and the basic threshold should be appropriately relaxed; liquid oxygen and liquid nitrogen are next; LNG is in the middle.

[0058] Based on historical data from K consecutive time windows preceding the current time window, where K ≥ 3 (e.g., K = 5), calculate the historical volatility factor λ. hist : ; In the formula, λ hist σ is the historical volatility factor; current The standard deviation of the current sliding window, in kPa / h; median(σ hist ) is the median of the standard deviations of the historical K sliding windows, in kPa / h. Using the median instead of the mean can reduce the impact of outliers on the estimation of historical volatility.

[0059] Historical volatility factor λ hist The range is 0.5 to 2.0, and a value greater than 1 indicates that the current fluctuation is higher than the historical median.

[0060] Calculate the adaptive steady-state determination threshold: ; In the formula, θfinal θ is the adaptive steady-state determination threshold. base The basic steady-state determination threshold, λ hist Historical volatility factor; The safety factor ranges from 0.8 to 1.2.

[0061] Safety factor Its function is to allow users to adjust the severity of the judgment according to the test accuracy requirements: When the value is large (e.g., 1.2), the threshold is relaxed, making it easier to enter a steady state, but this may sacrifice some accuracy. Smaller values ​​(e.g., 0.8) tighten the threshold, more rigorously ensuring steady state, but may prolong the waiting time. Safety factor The default value is 1.0.

[0062] The advantage of using an adaptive steady-state threshold is that when the gas cylinder tends to stabilize, σ current Gradually decrease, λ hist <1, θ final The threshold is correspondingly reduced (tightened), making the judgment conditions automatically more stringent, ensuring that it is only triggered when a deep steady state is truly reached; conversely, in the early stages of testing or when subjected to slight disturbances, the fluctuations are larger, and λ... hist >1, θ final Increase (relax the threshold) to avoid misjudging non-convergence due to normal fluctuations.

[0063] 5. Multidimensional feature extraction and trajectory construction Constructing multidimensional feature vectors This is a four-dimensional feature space, where r(t) reflects the rate of pressure change. Reflects changes in gas phase temperature. CV reflects the degree of environmental interference. r (t) reflects the relative fluctuation intensity.

[0064] Because the dimensions and numerical ranges of each component are different (e.g., the dimension of r is kPa / h, the dimension of T' is ℃ / min, CV...), r Since the vector is dimensionless, directly calculating the Euclidean distance and curvature will introduce bias due to the dimensionality of the vector. Therefore, before performing trajectory analysis, the components of the multidimensional feature vector are first normalized, mapping each component to the [0,1] interval: ; In the formula, The j-th eigencomponent after normalization, F j For the original j-th dimension feature component, min(F) j ) represents the minimum value of the j-th eigencomponent, max(F) j ) represents the maximum value of the j-th eigencomponent.

[0065] Where j represents the j-th component of the eigenvector, min(F j ) and max(F j It can be preset based on historical data or theoretical range, or it can be dynamically updated during the test.

[0066] Principal component analysis (PCA) is used to reduce the dimensionality of multidimensional eigenvectors by projecting the four-dimensional features into a two- or three-dimensional space before further analysis. PCA dimensionality reduction can reduce computational complexity while preserving key information and eliminating correlations between components.

[0067] The normalized feature vectors are arranged in chronological order, and a feature point is obtained at each window step (e.g., 5 minutes). The consecutive feature points are connected by line segments to form the state evolution trajectory.

[0068] Calculate the trajectory curvature and trajectory offset of the state evolution trajectory within a continuous time window: ; In the formula, For trajectory curvature, The first time derivative of the eigenvector. The second-order time derivative of the eigenvector; d represents the trajectory offset; d is the dimension of the feature vector. The normalized eigenvector at time t The j-th component; Reference equilibrium point vector The j-th component.

[0069] Reference equilibrium point The centroid of the 10 most recent feature points or a pre-defined steady-state center is used. Trajectory curvature. This reflects the abruptness of the state evolution transition: in the non-steady-state phase, the eigenvector changes rapidly, and the trajectory curvature is large; after entering the steady state, the eigenvector tends to stabilize, the trajectory approaches a straight line, and the curvature approaches zero. Trajectory offset. It reflects the degree to which the current state deviates from the equilibrium position. In steady state, it should fluctuate slightly around the equilibrium point.

[0070] First time derivative Second-order time derivative Approximate calculations were performed using the central difference method: , ; In the formula, This is the number of the current discrete time point. In order to be in Multidimensional feature vector at time step, The first time derivative of the eigenvector. The second-order time derivative of the eigenvector; The time step between two adjacent discrete moments; This is the feature vector for the next time step. This is the feature vector from the previous time step.

[0071] 6. Comprehensive judgment of steady-state stage When the pressure stability condition, temperature stability condition, and trajectory convergence condition are met simultaneously, the cryogenic adiabatic cylinder is determined to have entered the steady-state stage.

[0072] (1) Pressure stability condition: The coefficient of variation of the pressure growth rate sequence within a continuous time window is less than the adaptive steady-state determination threshold θ. final And the absolute value of the first-order difference of the pressure growth rate is less than the preset acceleration threshold ε r (e.g., ε) r =0.01 kPa / h²). A first-order difference condition is added to prevent the pressure growth rate from changing slowly and monotonically even though the coefficient of variation of the pressure growth rate is small.

[0073] (2) Temperature stability condition: The absolute value of each value in the gas phase temperature change rate sequence is less than the first temperature change rate threshold ε for three or more consecutive time windows. Tg (e.g., ε) r =0.05℃ / min), and the absolute value of each value in the ambient temperature change rate sequence is less than the second temperature change rate threshold ε for three consecutive time windows. Ta (e.g., ε) r =0.1℃ / min). The ambient temperature conditions ensured that external interference was basically eliminated.

[0074] 3. Trajectory Convergence Condition: The curvature κ of the evolution trajectory of the multidimensional feature vector composed of the pressure growth rate, the gas phase temperature change rate, and the ambient temperature change rate in the feature space is less than a preset curvature threshold κ. th (e.g., κ) th =0.02), and the trajectory offset δ is less than the preset offset threshold δ. th (e.g., δ) th =0.15), lasting for at least two consecutive time windows.

[0075] The three conditions described above characterize the stability of the thermodynamic processes inside the gas cylinder from different dimensions: the pressure stability condition focuses on fluctuations in directly measurable quantities; the temperature stability condition focuses on sources of disturbance; and the trajectory convergence condition focuses on the overall evolution trend of the system in high-dimensional space. The combined determination of these three conditions significantly improves the accuracy and robustness of steady-state identification.

[0076] 7. Multi-level anomaly identification and graded response In actual testing, various abnormal disturbances are unavoidable. These abnormal disturbances are categorized into three levels, and differentiated response strategies are adopted accordingly.

[0077] (1) Define three types of anomalous disturbances: Transient jump anomaly: Pressure data at a single sampling point exceeds the preset jump threshold ΔP. th Mutations (such as ΔP) th =3σ normal , where σ normal The fluctuation range is within the normal standard deviation, and the value returns to normal within the next 1 to 2 sampling periods. The causes of the anomaly are: transient noise in the data acquisition system, minor impact to the gas cylinder, or electronic interference, etc.

[0078] Continuous drift anomaly: The pressure increase rate exhibits a monotonically increasing or monotonically decreasing trend over multiple consecutive time windows (e.g., 5 windows), and the trend intensity exceeds the preset drift detection threshold (e.g., 0.01 kPa / h). 2 Trend detection can be performed using linear regression: for the window sequence (r1, r2, ..., r... m The fitted line is r = a·t + b, where r is the pressure increase rate, t is time (independent variable), a is the slope of the fitted line, and b is the intercept of the fitted line. If the slope |a| > 0.01 kPa / h 2 If this monotonic trend persists across multiple consecutive time windows (e.g., 5 windows), it is considered a persistent drift. The causes of this anomaly include: sensor zero-point drift, slow gas cylinder leakage, or continuous unidirectional changes in ambient temperature. This linear fitting method is simple and effective, enabling early identification of systemic problems such as sensor drift or slow gas cylinder leakage, avoiding misjudging a non-steady-state condition as a steady-state one.

[0079] An abnormal environmental step: The ambient temperature changes beyond the preset environmental step threshold ΔT within a short period of time (e.g., within 5 minutes). a,th (e.g., 2℃), and the change range exceeds three standard deviations from the rate of change of ambient temperature. The causes of the abnormality are: starting or stopping the air conditioning system, sudden opening of doors and windows, or sudden weather changes, etc.

[0080] (2) Implement a graded response based on the type of abnormal disturbance: For transient anomalies: the steady-state determination process is not paused; instead, a data repair algorithm is used to remove and replace outliers through interpolation. The data repair algorithm can be linear interpolation, cubic spline interpolation, or Kalman filtering.

[0081] For persistent drift anomalies: Pause the steady-state determination process and set the initial validity label for the current time window to invalid. The system continues monitoring, and after no persistent drift is detected for three or more consecutive time windows, the steady-state evolution analysis is re-executed. If the persistent drift lasts for more than 30 minutes, it is recommended to issue an alarm to prompt operators to check the sensor or gas cylinder status.

[0082] For environmental step anomalies: Pause the steady-state determination process and reset the historical data cache of the sliding window (clear the data of the most recent N windows, where N is the number of windows corresponding to the longest time window). After the ambient temperature stabilizes again (e.g., the rate of change of ambient temperature is less than 0.5℃ / h for 30 consecutive minutes), resume data accumulation and perform steady-state evolution analysis. This reset strategy avoids misjudgment caused by mixing data before and after the disturbance.

[0083] For example, during the settling process of a cryogenic insulated gas cylinder, the laboratory air conditioning system suddenly malfunctioned, causing the ambient temperature to rise from 20°C to 23°C within 3 minutes (a step change of 3°C). If the rate of change in ambient temperature was detected to be >0.5°C / min and the magnitude of the change exceeded 2°C, it was determined to be an environmental step anomaly. A tiered response was then executed: the steady-state determination process was immediately paused; the historical data cache of the sliding window was reset (clearing the data of the most recent 12 windows, corresponding to 2 hours); and a message was displayed on the screen: "A sudden change in ambient temperature detected, determination paused, waiting for environmental stabilization." After waiting 30 minutes, the ambient temperature stabilized within the range of 23±0.3°C, with a change rate <0.5°C / h, and the steady-state evolution analysis automatically resumed, accumulating data from scratch and re-determining the status. Finally, the steady-state stage was determined at the 16th hour.

[0084] The core advantage of the tiered response mechanism lies in its ability to: for repairable transient disturbances, maintain the testing process without interruption, performing only localized data cleaning; for persistent anomalies, promptly pause and resume testing; and for sudden environmental changes, completely reset the system to avoid historical data contamination. This differentiated approach significantly improves the system's robustness. This mechanism ensures that sudden environmental changes do not lead to erroneous decisions based on mixed data (before and after the change).

[0085] 8. Review before test launch Once the steady-state phase is confirmed, a dual verification is performed before the static evaporation rate test is automatically started to ensure the reliability of the judgment results.

[0086] First layer of verification: Duration verification. Verify whether the duration of the steady-state phase exceeds a preset minimum duration threshold T. min Tmin The value range is 0.5 to 2 hours, for example, a value of T. min= One hour. This duration-based verification helps avoid false positives caused by data accidentally meeting steady-state conditions.

[0087] Second verification: Deviation verification. Verifying the mean pressure increase rate μ during the steady-state phase. r，stable The average pressure growth rate μ compared to the previous stage (the last time window before entering the steady state stage) r，prev Is the relative deviation less than the preset verification deviation threshold η? th η th The value range is 5% to 10%, and the value is η. th =8. Deviation verification ensures that there is no significant jump in the rate of pressure change before and after entering steady state, thus guaranteeing the continuity of steady state.

[0088] If both of the above verifications pass, the steady-state phase is confirmed to be valid, a static evaporation rate test start signal is generated, and the static evaporation rate test process is triggered; otherwise, the steady-state evolution analysis continues until the verification passes.

[0089] It should be noted that the test start signal can be a digital output, an analog output, or a communication command. This invention is only responsible for determining the steady-state test conditions and starting the test, and does not limit the specific method for subsequent static evaporation rate testing. This invention belongs to a passive monitoring method under closed static conditions. After determining the steady-state conditions, it can seamlessly connect with various existing static evaporation rate testing methods to form a complete detection process. For example, the method for determining steady-state conditions provided by this invention can be used first, and then the rapid testing method can be used to measure the evaporation rate, thereby achieving full-process optimization of rapid testing.

[0090] Please see Figure 7 This embodiment also provides a steady-state determination system for static evaporation rate testing of cryogenic insulated gas cylinders, used to implement the above-mentioned steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders. The system includes: The data acquisition module 100 is used to allow the cryogenic insulated gas cylinder to stand after the cryogenic medium is filled, and to continuously collect the internal pressure data, gas phase space temperature data and ambient temperature data of the gas cylinder at a preset sampling period. The feature parameter calculation module 200 is connected to the data acquisition module 100 and is used to calculate the pressure growth rate sequence based on the pressure data at adjacent sampling times, calculate the gas phase temperature change rate sequence based on the gas phase space temperature data at adjacent sampling times, and calculate the ambient temperature change rate sequence based on the ambient temperature data at adjacent sampling times. The multi-scale analysis module 300, connected to the characteristic parameter calculation module 200, is used to perform parallel sliding window analysis on the pressure growth rate sequence using short and long time windows to determine whether the fluctuation of the pressure growth rate tends to converge at both the local and global scales. The steady-state judgment module 400 is connected to the characteristic parameter calculation module 200 and the multi-scale analysis module 300, respectively, and is used to determine whether the cryogenic adiabatic cylinder has entered the steady-state stage.

[0091] Example 1: Steady-state determination test of commonly used liquid nitrogen adiabatic gas cylinders The following example illustrates the steady-state determination before conducting a static evaporation rate test on a 450L liquid nitrogen insulated cylinder. The experimental conditions are as follows: filling medium: liquid nitrogen (purity 99.999%), initial filling rate: 92.5%; ambient temperature: 20±2℃, ambient air pressure: 101.3±0.2 kPa, ambient airflow: natural convection, no forced wind, wind speed <0.2m / s.

[0092] 1. Data Collection After liquid nitrogen filling is completed, the cryogenic insulated gas cylinder is placed statically on the test platform with rubber shock-absorbing pads at the bottom. A high-precision pressure sensor with a range of 0-2 MPa, an accuracy of 0.1% FS, and an output of 4-20 mA is installed near the cylinder's discharge port. A platinum resistance temperature sensor is placed in the gas phase space inside the cylinder; and an ambient temperature sensor is placed 1 meter away from the cylinder at a height of 1.5 meters, with a radiation shield installed. The sampling period is set to 10 seconds, continuously collecting pressure, gas phase temperature, and ambient temperature data, which are stored in the database in real time.

[0093] 2. Calculation of characteristic parameters The pressure increase rate is calculated every 10 seconds and converted into an hourly rate of change. The calculation formula is as follows: ; For example, if the pressure increases by 0.0015 kPa from its initial value within 10 seconds, then the rate of pressure increase is: r = (0.0015 / 10)×3600 = 0.54 kPa / h.

[0094] Please see Figure 2 and Figure 3 The pressure increases rapidly during the initial filling stage; as the cylinder cools, the pressure gradually decreases and stabilizes.

[0095] Simultaneously calculate the rate of change of vapor phase temperature. and the rate of change of ambient temperature The rate of change per minute is calculated by taking the average value every 10 seconds (i.e., averaging over 6 sampling points), and the unit is ℃ / min. Calculation formula: ; ; Each sampling point (i.e., 5 minutes) constitutes a basic time window. The mean, standard deviation, and coefficient of variation of the pressure increase rate within this window are calculated: ; Two hours after filling, μ r About 1.2 kPa / h, σ r Approximately 0.2 kPa / h, CV r ≈0.167. At the 12th hour, μ r reduced to 0.35 kPa / h, σ r Reduced to 0.025 kPa / h, CV r ≈0.071.

[0096] 3. Multi-scale sliding window analysis Please see Figure 4 Set a short time window W s =10 minutes (corresponding to 60 sampling points), long window W L = 60 minutes (corresponding to 360 sampling points). The sliding step is set to 5 minutes (30 sampling points), that is, the judgment result is updated every 5 minutes.

[0097] Calculate the CV in each of the two windows. r Values. θ1 = 0.05 (i.e., local fluctuations do not exceed 5% of the mean), θ2 = 0.10 (i.e., overall fluctuations do not exceed 10% of the mean). These thresholds are determined based on historical test data for this type of gas cylinder: in tests confirming a steady state, CV... r (W s The mean of ) is 0.04, and the standard deviation is 0.01; CV r (W L The mean of the values ​​is 0.07 and the standard deviation is 0.015. We take the mean + 2σ as the threshold, which is 0.04 + 0.02 = 0.06 (we actually take 0.05 to leave a margin) and 0.07 + 0.03 = 0.10.

[0098] 4. Dynamic adaptive threshold generation Cylinder parameters: nominal volume 450L, design pressure 1.6MPa, medium is liquid nitrogen, insulation structure is high vacuum multilayer insulation (vacuum degree <5×10⁻⁶). -3 Pa (30 layers). The basic steady-state determination threshold θ is obtained through a pre-trained regression model (trained based on 50 sets of historical test data). base= 0.08.

[0099] We set K=5, which means we take the standard deviation (σ) of the five time windows preceding the current window. hist1 , σ hist2 , σ hist3 , σ hist4 ,σ hist5 ), calculate the median(σ) hist Taking the determination of the 12th hour as an example: the standard deviations of the previous 5 windows were 2.38, 2.35, 2.31, 2.28, and 2.25 kPa / h, respectively, with a median of 2.31 kPa / h. The standard deviation σ of the current window... current =2.21 kPa / h, then λ hist = 2.21 / 2.31≈0.957. Take a safety factor α=1.0 (standard accuracy requirement).

[0100] Calculate the adaptive steady-state determination threshold: .

[0101] It can be seen that as the gas cylinder tends to stabilize, λ hist As the threshold gradually decreases to less than 1, it automatically tightens, and the judgment conditions become more stringent.

[0102] 5. Multidimensional feature extraction and trajectory construction Constructing four-dimensional feature vectors First, normalization is performed. Based on historical data, the typical range of each component is as follows: r(t): 0~3 kPa / h → Mapped to [0,1]; T'g(t): -0.5~0.5 ℃ / min → Mapped to [0,1]; T'a(t): -0.3~0.3 ℃ / min → Mapped to [0,1]; CV r : 0~0.3 → Map to [0,1].

[0103] Taking r as an example, the normalization formula is: ; A normalized feature point is calculated every 5 minutes, and the continuous feature points are projected onto a four-dimensional feature space (the first three dimensions are selected for visualization in actual display) to form a state evolution trajectory. Local weighted regression smoothing is used to remove noise from the trajectory.

[0104] Calculate the trajectory curvature κ and trajectory offset δ. In the normalized feature space, set a curvature threshold κ. th =0.02, offset threshold δ th= 0.15. At the 8th hour, the trajectory curvature κ is approximately 0.035 and the offset δ is approximately 0.22; at the 12th hour, κ decreases to 0.018 and δ decreases to 0.12.

[0105] 6. Comprehensive judgment of steady-state stage Please see Figure 4 and Figure 5 The timing begins from the completion of filling the cryogenic insulation gas cylinder, and a steady-state determination is performed every 5 minutes.

[0106] (1) 8-10 hours after filling Pressure stability condition: CV r (W s =0.072, CV r (W L ) = 0.095, θ final =0.077. Where, CV r (W s )<θ final Established, but CV r (W L )>θ final Overall, the fluctuations are still too large, and the conditions are not met.

[0107] Temperature stability conditions: The average absolute value of the gas phase temperature change rate is about 0.012℃ / min (less than 0.05℃ / min), and the average absolute value of the ambient temperature change rate is about 0.008℃ / min (less than 0.1℃ / min), and these conditions have been met for 4 hours.

[0108] Trajectory convergence condition: Trajectory curvature κ=0.031>0.02, which is not satisfied.

[0109] Overall assessment: The market has not yet reached a steady state; monitoring will continue.

[0110] (2) 12-14 hours after filling Pressure stability condition: CV r (W s =0.058, CV r (W L ) = 0.069, all less than θ final Furthermore, the absolute value of the first-order difference is 0.0012 kPa / h², which is less than 0.01 kPa / h², so the condition is met.

[0111] Temperature stability condition: Temperature stability has been continuously met for more than 2 hours.

[0112] Trajectory convergence conditions: trajectory curvature κ=0.018<0.02, offset δ=0.12<0.15, the conditions are met.

[0113] If all three conditions are met, the system is considered to have entered a steady state.

[0114] 7. Abnormal Disturbance Identification and Graded Response During the static period of the cryogenic insulated gas cylinder, continuous monitoring for abnormalities was conducted, and no abnormal disturbances occurred.

[0115] 8. Review before test launch Once the steady-state phase is confirmed (after the 12-hour judgment is passed), the system begins dual verification.

[0116] (1) Duration verification: After the steady state was confirmed at the 12th hour, monitoring continued until the 14th hour. The steady state duration reached 2 hours, exceeding T. min =1 hour, verification passed.

[0117] (2) Deviation verification: mean pressure growth rate μ during steady-state phase (12-14 hours) r,stable =0.32 kPa / h, the average value of the previous period (10-12 hours) μ r,prev =0.35 kPa / h, relative deviation = |0.32-0.35| / 0.35≈8.6%, which is less than η. th =10%, verification passed.

[0118] After the verification is passed, a start command is sent to the static evaporation rate testing device. Upon receiving the start signal, the testing device begins the static evaporation rate test.

[0119] The entire process, from filling to test commencement, took approximately 14 hours. The static evaporation rate was calculated to be 1.28% / day after the test. In comparison, the traditional 48-hour settling method (using the same cylinder, retested after a one-week interval) yielded 1.21% / day, a relative deviation of approximately 5.8%. Considering test repeatability—that is, the slight differences in cylinder condition and environmental conditions between the two tests—this deviation is within the engineering tolerance range (typically required to be <10%).

[0120] Please see Figure 6 The other two gas cylinders were also tested: cylinder A2 was a 450L cylinder of the same model, and cylinder B1 was a 175L cylinder. The test results are summarized in Table 1.

[0121] Table 1 Comparison of test results between the two methods It can be seen that the average relative deviation of the method of the present invention is about 4.3%, and the average determination time is about 12.8 hours, which is less than the 48 hours of the traditional method.

[0122] Example 2: Adaptability Verification of Cryogenic Insulation Cylinders of Different Specifications 1. Select three cryogenic insulated gas cylinders with different volumes and media types, and conduct tests on them respectively.

[0123] 100L liquid oxygen cylinder (small, medium liquid oxygen): high vacuum multi-layer insulation, nominal working pressure 2.0MPa.

[0124] 450L liquid nitrogen cylinder (medium size, medium: liquid nitrogen): high vacuum multilayer insulation, nominal working pressure 1.6MPa.

[0125] 1000L LNG cylinder (large, medium: liquefied natural gas): high vacuum multi-layer insulation, nominal working pressure 1.8MPa.

[0126] 2. Each type of gas cylinder was tested using the following methods. (1) Traditional 48-hour settling method; (2) The method of the present invention (default parameter, α=1.0); (3) The method of the present invention (tightening parameter, α=0.9); (4) The method of the present invention (relaxed parameters, α=1.1).

[0127] Compare the static evaporation rate results and judgment time for each test; see Table 2 for test results.

[0128] Table 2. Adaptability test results for different gas cylinders 3. Test data analysis For these three types of gas cylinders, the test results of the method of this invention deviated from those of the traditional method by less than 6%, and the judgment time was shortened by approximately 39.5 hours (100L gas cylinder), 34 hours (450L gas cylinder), and 26 hours (1000L gas cylinder), respectively. The reduction in judgment time was more significant for smaller gas cylinders because smaller gas cylinders reach thermal equilibrium more quickly.

[0129] The severity of the judgment can be changed by adjusting the safety factor α: When α=0.9, the deviation is smaller (<2%), but the judgment time is longer; When α=1.1, the deviation increases (7-9%), but the judgment time is shortened.

[0130] The appropriate α value can be selected based on the actual accuracy requirements. For initial testing or type testing, use α=0.9 (more stringent); for periodic testing, use α=1.0 (balanced); and for rapid screening, use α=1.1 (faster).

[0131] For LNG (boiling point -162℃, lower than liquid nitrogen's -196℃), the difference between its saturation temperature and the ambient temperature is greater, resulting in a higher pressure increase rate, but the fluctuations are relatively stable. A dynamic threshold mechanism can automatically adapt without requiring manual parameter adjustments.

[0132] 4. Test Conclusion The steady-state determination method for static evaporation rate testing of cryogenic insulated gas cylinders provided by this invention has good universality and can be applied to cryogenic insulated gas cylinders of different volumes (100L~1000L) and different media (such as liquid oxygen, liquid nitrogen, or LNG). Furthermore, by adjusting the safety factor α, different accuracy-time balance requirements can be met. Significant progress has been made in accuracy, efficiency, robustness, automation, and compatibility, demonstrating broad prospects for industrial applications. Experiments show that the steady-state determination accuracy of this method exceeds 98%, while methods based on a single fixed threshold only achieve around 85%.

[0133] The embodiments of the present invention have the following beneficial effects: 1. High accuracy: By extracting multi-dimensional features (pressure, gas phase temperature, and ambient temperature) and using multi-scale sliding window (short window + long window) analysis, the system can more accurately identify the true timing when the gas cylinder enters steady state. In particular, the introduction of trajectory convergence conditions captures steady-state characteristics from the overall perspective of system evolution, avoiding misjudgments due to the accidental fulfillment of conditions by a single parameter.

[0134] 2. Strong environmental adaptability: The dynamic adaptive threshold mechanism allows the judgment conditions to be automatically adjusted according to the cylinder specifications, medium type, and real-time data fluctuations, making it suitable for cylinders of different volumes from 100L to 1000L, as well as different media such as liquid nitrogen, liquid oxygen, and LNG. In contrast, the traditional fixed threshold method requires repeated parameter adjustments for different cylinders.

[0135] 3. High Robustness: The three-level anomaly identification and graded response mechanism effectively distinguishes between transient disturbances, persistent drift, and environmental mutations, and adopts differentiated handling strategies for different types of anomalies. Transient disturbances only require data repair without pausing the judgment process, avoiding test interruption; persistent drift and environmental mutations pause the judgment process and take corresponding recovery measures. This design enables the system to operate stably in noisy industrial environments without misjudging or crashing due to occasional interference.

[0136] 4. High Reliability: Dual verification before startup ensures that the steady-state phase has been established stably. Duration verification avoids false positives caused by data occasionally meeting the conditions; deviation verification ensures the continuity of data between the steady-state phase and the previous phase, without any jumps. Experiments show that after starting the test with dual verification, the deviation from the traditional 48-hour settling method can be controlled within 3%.

[0137] 5. Short test preparation time: While ensuring the accuracy of test results, the preparation time for static evaporation rate testing is reduced from the traditional 48 hours to 12-18 hours. For batch testing scenarios, the time saving effect is particularly significant. Based on testing 300 gas cylinders per year, approximately 9,000 hours of testing time can be saved, while reducing liquid nitrogen consumption by approximately 15 tons.

[0138] 6. Less reliance on human experience: It achieves full automation and intelligence in steady-state determination, and the consistency of test results between different operators is reduced from ±15% in traditional manual judgment to within ±3%, which significantly improves the objectivity and reproducibility of test results.

[0139] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for determining the steady-state stability of static evaporation rate testing of a cryogenic adiabatic gas cylinder, characterized in that, include: After the cryogenic medium is filled, the cryogenic insulated gas cylinder is left to stand, and the internal pressure data, gas phase space temperature data and ambient temperature data are continuously collected at a preset sampling period. Based on pressure data from adjacent sampling times, the pressure change per unit time is calculated to form a pressure growth rate sequence. Based on the gas phase space temperature data at adjacent sampling times, the change in gas phase temperature per unit time is calculated to form a sequence of gas phase temperature change rates. Based on the ambient temperature data at adjacent sampling times, the change in ambient temperature per unit time is calculated to form an ambient temperature change rate sequence. The pressure growth rate sequence is subjected to parallel sliding window analysis using short and long time windows to determine whether the fluctuation of the pressure growth rate tends to converge within the continuous time window. When the pressure stability condition, temperature stability condition, and trajectory convergence condition are met simultaneously, the cryogenic adiabatic cylinder is determined to have entered the steady state stage. The pressure stability condition is as follows: the fluctuation of the pressure growth rate tends to converge within a continuous time window, and the coefficient of variation of the pressure growth rate within the continuous time window is less than the adaptive steady-state determination threshold. The temperature stability condition is as follows: the absolute value of each value in the gas phase temperature change rate sequence is less than the first temperature change rate threshold for three or more consecutive time windows, and the absolute value of each value in the ambient temperature change rate sequence is less than the second temperature change rate threshold for three or more consecutive time windows. The trajectory convergence condition is as follows: the curvature of the evolution trajectory of the multidimensional feature vector composed of the pressure growth rate, the gas phase temperature change rate and the ambient temperature change rate in the feature space is less than a preset curvature threshold, and the trajectory offset is less than a preset offset threshold, for at least two consecutive time windows. The generation of the adaptive steady-state determination threshold includes: Obtain the nominal volume, design pressure rating, cryogenic medium type, and insulation structure of the cryogenic insulated gas cylinder, and generate the basic steady-state judgment threshold. Based on historical data from K consecutive time windows prior to the current time window, a historical volatility factor is calculated, which is equal to the ratio of the current sliding window standard deviation to the median of the historical sliding window standard deviations. The adaptive steady-state determination threshold is obtained by multiplying the basic steady-state determination threshold by the historical fluctuation amplitude factor and the preset safety factor, wherein the value of the safety factor ranges from 0.8 to 1.

2. The method also includes a multi-level anomaly identification and graded response step: Identify the types of abnormal disturbances, including transient jump anomalies, continuous drift anomalies, and environmental step anomalies; Implement a tiered response based on the type of abnormal disturbance: For transient jump anomalies, the steady-state determination process is not paused; instead, a data repair algorithm is used to remove and replace the anomalies. For persistent drift anomalies, the steady-state determination process is paused, and the steady-state evolution analysis is re-executed after no persistent drift is detected for three or more consecutive time windows. For environmental step anomalies, the steady-state determination process is paused, the historical data cache of the sliding window is reset, and the steady-state evolution analysis is re-executed after the environmental temperature stabilizes again.

2. The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to claim 1, characterized in that, The preset sampling period is adaptively set according to the gas cylinder volume, and the sampling period range is 1 to 60 seconds.

3. The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to claim 1, characterized in that, The short time window has a length of 5 to 30 minutes, and the long time window has a length of 30 to 120 minutes. When the coefficient of variation of the pressure growth rate within the short time window is less than a first local threshold, and the coefficient of variation of the pressure growth rate within the long time window is less than a second local threshold, it is determined that the pressure growth rate tends to converge on both the local and global scales, wherein the first local threshold is less than the second local threshold.

4. The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to claim 1, characterized in that, The multidimensional feature vectors are subjected to dimensionality reduction using principal component analysis, which projects the high-dimensional features into a low-dimensional space before calculating the trajectory curvature and trajectory offset.

5. The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to claim 1, characterized in that, The formula for calculating the pressure change per unit time to form a pressure growth rate sequence is as follows: ； In the formula, r for The rate of pressure increase over time, expressed in kPa / h. for The pressure value at any moment, for The pressure value at any given moment; The sampling period is in seconds; 3600 is a time unit conversion factor, which converts the rate of change per second to the rate of change per hour. The formula for calculating the change in gas phase temperature per unit time to form a sequence of gas phase temperature change rates is as follows: ； In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is expressed in seconds. 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute. The formula for calculating the change in ambient temperature per unit time, forming a sequence of ambient temperature change rates, is as follows: ； In the formula, For t i The rate of change of gas phase temperature at any given time, expressed in °C / min; For t i The temperature of the gas phase space at any given time. For t i-1 The temperature of the gas phase space at any given moment; The sampling period is expressed in seconds. 60 is a time unit conversion factor, which converts the rate of change per second to the rate of change per minute. The formulas for calculating the mean, standard deviation, and coefficient of variation of the pressure increase rate within a continuous time window are as follows: ； In the formula, μ r r represents the average rate of pressure increase within the sliding window, expressed in kPa / h. j σ represents the j-th pressure increase rate within the sliding window, in kPa / h; n is the number of sampling points within the sliding window; r CV represents the standard deviation of the pressure increase rate within the sliding window, in kPa / h. r is the coefficient of variation of the pressure increase rate.

6. The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to claim 1, characterized in that, After the step of determining that the cryogenic insulated gas cylinder has entered the steady-state stage, the method further includes: Verify whether the duration of the steady-state phase exceeds a preset minimum duration threshold, wherein the minimum duration threshold ranges from 0.5 to 2 hours; The test verifies whether the relative deviation between the average pressure growth rate during the steady-state phase and the average pressure growth rate in the previous phase is less than a preset verification deviation threshold, wherein the value of the verification deviation threshold ranges from 5% to 10%. If all the above verifications pass, the steady-state phase is confirmed to be effective, and a static evaporation rate test start signal is generated; otherwise, the steady-state evolution analysis continues until the verification passes.

7. The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to claim 1, characterized in that, The instantaneous jump anomaly is defined as: a sudden change in pressure data at a single sampling point that exceeds a preset jump threshold, and which returns to normal within the next 1 to 2 sampling periods; The continuous drift anomaly is defined as follows: the pressure growth rate shows a monotonically increasing or monotonically decreasing trend within multiple consecutive time windows, and the trend intensity exceeds the preset drift detection threshold. The environmental step anomaly is defined as: the ambient temperature changes beyond a preset environmental step threshold within a short period of time, and the magnitude of the change exceeds three times the standard deviation of the rate of change of ambient temperature.

8. A steady-state determination system for static evaporation rate testing of cryogenic insulated gas cylinders, characterized in that, The method for determining the steady-state stability of static evaporation rate testing of cryogenic insulated gas cylinders according to any one of claims 1 to 7 includes: The data acquisition module is used to allow the cryogenic insulated gas cylinder to stand after it has been filled with cryogenic medium, and to continuously collect internal pressure data, gas phase space temperature data and ambient temperature data of the gas cylinder at a preset sampling period. The feature parameter calculation module, connected to the data acquisition module, is used to calculate the pressure growth rate sequence based on the pressure data at adjacent sampling times, calculate the gas phase temperature change rate sequence based on the gas phase space temperature data at adjacent sampling times, and calculate the ambient temperature change rate sequence based on the ambient temperature data at adjacent sampling times. The multi-scale analysis module, connected to the feature parameter calculation module, is used to perform parallel sliding window analysis on the pressure growth rate sequence using short and long time windows to determine whether the fluctuation of the pressure growth rate tends to converge at both the local and global scales. The steady-state judgment module is connected to the feature parameter calculation module and the multi-scale analysis module respectively, and is used to determine whether the cryogenic adiabatic cylinder has entered the steady-state stage.

Images